This is a novel approach to identify how visual content can be used in the field of marketing with respect to the demographic factors of the users. In this research, a large amount of data was collected from users of different age, racial, and ethnic groups, along with their selection of clothing items.
The objective of this research is to build a classification model through which marketers or stakeholders in the garment sector can identify a potential customer base for a newly created design based on an image.
A multi-label classification model was built using image data and textual data, specifically the demographic data associated with the selected clothing images of the users.
When an image is given to the model, it outputs the potential demographic factors through which this garment can gain popularity.
This research was conducted in the Sri Lankan context and is the first approach of its kind, particularly combining the selection of images with the demographic factors of those who chose the clothing items.
For an in-depth exploration of this approach and detailed findings, refer to my thesis: Visual Content Usage for Marketing on Demography